Conversational Document Prediction to Assist Customer Care Agents
This work addresses a specific problem for customer care agents by improving document retrieval efficiency, but it is incremental as it builds on existing IR and DL methods.
The paper tackles the problem of predicting relevant documents for customer care agents to use during conversations, introducing a new public dataset and evaluating deep learning and information retrieval models. The result shows that a hybrid IR+DL approach offers the best performance, with analysis of inference time complexity.
A frequent pattern in customer care conversations is the agents responding with appropriate webpage URLs that address users' needs. We study the task of predicting the documents that customer care agents can use to facilitate users' needs. We also introduce a new public dataset which supports the aforementioned problem. Using this dataset and two others, we investigate state-of-the art deep learning (DL) and information retrieval (IR) models for the task. Additionally, we analyze the practicality of such systems in terms of inference time complexity. Our show that an hybrid IR+DL approach provides the best of both worlds.